A quality metric to assess pulsewise-coil displacements during TMS

124 

Abstract Submission 

Ole Numssen¹, Kathleen Williams¹, Sandra Martin¹, Thomas Knösche², Gesa Hartwigsen³

¹Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Saxony, ²Max Planck Institute, Leipzig, Saxony, ³Leipzig University, Leipzig, Saxony

Ole Numssen  
Max Planck Institute for Human Cognitive and Brain Sciences
Leipzig, Saxony

Kathleen Williams  
Max Planck Institute for Human Cognitive and Brain Sciences
Leipzig, Saxony

Sandra Martin  
Max Planck Institute for Human Cognitive and Brain Sciences
Leipzig, Saxony

Thomas Knösche  
Max Planck Institute
Leipzig, Saxony

Gesa Hartwigsen  
Leipzig University
Leipzig, Saxony

Transcranial Magnetic Stimulation (TMS) has emerged as a powerful non-invasive brain stimulation technique with applications in both clinical and research settings. Recently, the field has seen a surge in methodological rigor [1,2], yet a notable gap persists - the absence of a standardized quality metric to thoroughly assess and report pulse-by-pulse placement accuracy in TMS applications. Especially in manually-guided TMS, where experimenters compensate for movements, coil displacements significantly impact stimulation exposure, potentially hindering therapeutic or scientific outcomes. Despite extensive use of neuronavigation systems, there is currently no straightforward metric to quantify this critical factor. Unlike other neuroscientific modalities, such as functional magnetic resonance imaging (fMRI), where motion quantification is firmly integrated [3], TMS lacks a standardized method to quantify TMS coil displacement during experiments.
Here, we introduce a novel metric for TMS coil displacement: pulsewise coil displacement (PCD). PCD combines three positional (x, y, and z) and three rotational parameters (α, β, γ) for each TMS pulse into one compound metric. This metric, PCD, offers a meaningful and straightforward assessment tool for both trial-to-trial stimulation accuracy and the overall quality of a TMS experiment. PCD fills a critical void, providing researchers and clinicians with means to evaluate and report the accuracy of TMS applications, contributing to enhanced methodological rigor and reporting standards in the field.

TMS coil placements are tracked by neuronavigation systems with six parameters defining its position and orientation [Fig. 1a]. To assess positional displacements in a meaningful manner, a coordinate transformation is applied to differentially quantify orthogonal and tangential coil movements. Rotational displacement of the coil was assessed by transforming roll and pitch displacements into positional displacements using a beam projection method. With this approach, the roll and pitch displacements are transformed into a positional displacement of the stimulation center at a specific skin-cortex distance. Yaw displacements are quantified separately, because effects of yaw displacements critically depend on the coil geometry. To provide one metric, yaw displacements are included in the PCD compound metric by sin(yaw) (Fig. 1b).
Validation: (1) For a large set of virtual TMS experiments (50,000 pulses) e-fields [4] at target and off-targets were extracted to analyze PCD's correlation with cortical stimulation exposure. (2) To assess PCD's correlation with online TMS effects we analyzed a dataset on primary motor cortex stimulation and motor evoked potentials (MEPs) [5]. (3) Finally, we used linear mixed models to identify PCD's potential to explain variance in cortical activity modulation after cTBS to the left and right inferior parietal lobe.

(1) PCD, consolidating information from all six displacement parameters (Fig. 2a), exhibited significant correlations with the induced e-fields at target (Fig. 2b) and off-target regions. (2) Validation against motor-evoked potentials (MEPs) demonstrated a similar correlation strength of r(PCD, MEP) and r(e-field, MEP) (Fig. 2c). (3) PCD explained variance in local cortical activity modulation. Specifically, PCD explained fractional amplitude of low frequency fluctuations (fALFF) [6] variance in stimulated regions only (Fig. 2c).

We present a novel metric to quantify subject- or TMS-coil-movement throughout a TMS experiment or therapeutic intervention in a pulse-by-pulse manner. The validation against physical and physiological effects underlines its capabilities to capture relevant variance of TMS effects stemming from experimental imperfections. We provide means for automated PCD quantification within our pyNIBS [5] python package. Potential applications of PCD include quality control, statistical model strengthening, and intervention monitoring.

Non-invasive Magnetic/TMS

TMS ¹

Methods Development

Motion Correction and Preprocessing ²

Informatics Other

Other - Quality assessment; Reporting; motion correction;